TGIF Transcription Factors Repress Acetyl Coa Metabolic Gene Expression and Promote Intestinal Tumor Growth

Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

TGIF transcription factors repress acetyl CoA metabolic gene expression and promote intestinal tumor growth

Anant Shah,1 Tiffany A. Melhuish,1 Todd E. Fox,2 Henry F. Frierson Jr,3 and David Wotton1 1Department of Biochemistry and Molecular Genetics, Center for Cell Signaling, University of Virginia, Charlottesville, Virginia 22908, USA; 2Department of Pharmacology, 3Department of Pathology, University of Virginia, Charlottesville, Virginia 22908, USA

Tgif1 (thymine–guanine-interacting factor 1) and Tgif2 repress gene expression by binding directly to DNA or interacting with transforming growth factor (TGF) β-responsive SMADs. Tgifs are essential for embryogenesis and may function in tumor progression. By analyzing both gain and loss of Tgif function in a well-established mouse model of intestinal cancer, we show that Tgifs promote adenoma growth in the context of mutant Apc (adenomatous polyposis coli). Despite the tumor-suppressive role of TGFβ signaling, transcriptome profiling of colon tumors suggests minimal effect of Tgifs on the TGFβ pathway. Instead, it appears that Tgifs, which are up-regulated in Apc mutant colon tumors, contribute to reprogramming metabolic gene expression. Integrating gene expression data from colon tumors with other gene expression and chromatin-binding data identifies a set of direct Tgif target genes encoding proteins involved in acetyl CoA and pyruvate metabolism. Analysis of both tumor and nontumor tissues indicates that these genes are targets of Tgif repression in multiple settings, suggesting that this is a core Tgif function. We propose that Tgifs play an important role in regulating basic energy metabolism in normal cells, and that this function of Tgifs is amplified in some cancers. [Keywords: TGFβ; Tgif; Wnt; cancer; colorectal cancer; gene expression; metabolism; repressor] Supplemental material is available for this article. Received August 24, 2018; revised version accepted January 24, 2019.

Colorectal cancer (CRC) is among the most frequently di- sulting in phosphorylation and activation of the SMAD agnosed cancers and is the third leading cause of cancer transcription factors (Feng and Derynck 2005; Massagué deaths in adults (Miller et al. 2016). The adenomatous pol- et al. 2005). TGFβ, Activin, or Nodal signaling results pri- yposis coli (APC) gene, which is mutated in up to 80% of marily in activation of SMAD2 and SMAD3, which, in as- sporadic CRC (Kinzler and Vogelstein 1996; Segditsas and sociation with SMAD4, accumulate in the nucleus to Tomlinson 2006), encodes a scaffold-like protein that as- activate gene expression. TGFβ signaling is tumor suppres- sembles a complex of proteins including β-catenin, Axins, sive in many cancers, in part due to the antiproliferative ef- and GSK3β (Sancho et al. 2004; Clevers and Nusse 2012). fects of the pathway (Levy and Hill 2006; Massagué 2008). In the absence of Wnt ligands, β-catenin is phosphorylated Inactivating mutations in the TGFBR2 gene, encoding the and degraded. Wnt signaling prevents phosphorylation type II receptor, are found in around 25% of CRC (Marko- and destruction of β-catenin, allowing it to accumulate witz et al. 1995; Grady et al. 1999). Loss of heterozygosity and activate gene expression. Mutation or deletion of of a region of chromosome 18 that includes both SMAD2 APC results in constitutive β-catenin-dependent gene ac- and SMAD4 is seen in almost 70% of CRC but is less fre- tivation, primarily via interactions with the TCF/LEF quently observed in colorectal adenomas (Vogelstein transcription factors (Polakis 1995; Sancho et al. 2004). et al. 1988). Conditional inactivation of the mouse Tgfbr2 Inactivation of one allele of the mouse Apc gene by gene in combination with an Apc mutation results in pro- germ-line mutation (Moser et al. 1990; Fodde et al. 1994; gression to invasive adenocarcinoma, whereas homozy- Oshima et al. 1995) or by CRE-mediated excision of a gous Tgfbr2 deletion alone in the intestine has no effect loxP-flanked exon (Shibata et al. 1997; Colnot et al. (Muñoz et al. 2006). 2004; Hinoi et al. 2007) causes multiple adenomas, due TGIF1 (thymine–guanine-interacting factor) and the to sporadic inactivation of the remaining allele. closely related TGIF2 are homeodomain transcription In response to transforming growth factor β (TGFβ) ligands, the type I and type II receptors form a complex, re- © 2019 Shah et al. This article is distributed exclusively by Cold Spring Harbor Laboratory Press for the first six months after the full-issue publication date (see http://genesdev.cshlp.org/site/misc/terms.xhtml). After Corresponding author: [email protected] six months, it is available under a Creative Commons License (Attribu- Article published online ahead of print. Article and publication date are tion-NonCommercial 4.0 International), as described at http://creative- online at http://www.genesdev.org/cgi/doi/10.1101/gad.320127.118. commons.org/licenses/by-nc/4.0/.

GENES & DEVELOPMENT 33:1–15 Published by Cold Spring Harbor Laboratory Press; ISSN 0890-9369/19; www.genesdev.org 1 Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Shah et al. factors that are members of the TALE (three-amino-acid Wnt/β-catenin pathway, although this appeared to be in- loop extension) superfamily (Bertolino et al. 1995; Melhu- dependent of effects on Axins (Wang et al. 2017). Thus, ish et al. 2001; Hyman et al. 2003). Other TALE homeodo- Tgifs can promote tumorigenesis, but questions regarding main proteins include the Meis and Pbx families that mechanisms of action and potential overlapping roles of activate gene expression (Bürglin and Affolter 2016). In Tgif1 and Tgif2 remain. contrast, TGIF1 and TGIF2 are transcriptional repressors We used genetically engineered mouse models to that interact with multiple general corepressors, includ- address the function of Tgifs in intestinal epithelial neo- ing mSin3 and histone deacetylases (Wotton et al. plasms and to identify downstream Tgif target genes. 1999a; Melhuish et al. 2001). In addition, TGIF1 recruits Overexpression of TGIF1 in intestinal epithelial cells in- CtBP1/2 corepressors via a conserved interaction motif creased the size and number of adenomas in the small in- (Melhuish and Wotton 2000). Tgifs limit the response to testine, and deletion of Tgif1 and Tgif2 reduced tumor size TGFβ signaling by recruiting corepressors to the SMAD in both the small intestine and colon. Transcriptional pro- transcription factors (Wotton et al. 1999a; Melhuish filing of colon tumors from these mice revealed little ef- et al. 2001). In addition to SMAD interaction, other mech- fect of Tgifs on either Wnt/β-catenin or TGFβ signaling. anisms for TGFβ pathway inhibition have been suggested, Instead, we found that deleting Tgifs from colon tumors including promoting SMAD2 ubiquitylation and degrada- caused changes in expression of genes affecting multiple tion or preventing SMAD2 phosphorylation in response to metabolic pathways. Integrating these data with addition- TGFβ signaling (Seo et al. 2004, 2006). Loss-of-function al gene expression profiling results suggests that Tgifs play TGIF1 mutations are associated with holoprosencephaly a fundamental role in regulating energy metabolism, and (HPE), a severe genetic disease affecting forebrain develop- may contribute to the reprogramming of metabolic gene ment (Wotton and Taniguchi 2018). Mouse models of expression that occurs in CRC. Tgif1 and Tgif2 loss of function suggest they play a redundant, but essential role in early embryogenesis (Powers et al. 2010). Conditional mutants survive to mid-gestation Results with multiple developmental abnormalities, including Increased Tgif expression in colorectal tumors HPE (Taniguchi et al. 2012, 2017). Although developmental defects in embryos lacking Analysis of TCGA colorectal data showed elevated TGIF1 Tgif1 and Tgif2 can be partly rescued by reducing TGFβ and TGIF2 in adenocarcinomas (Fig. 1A). Similar results family signaling through mutation of Nodal (Powers were seen with additional CRC data sets, and in compari- et al. 2010; Taniguchi et al. 2012, 2017), transcriptome son of paired tumor and normal samples, TGIF1 expres- profiling of early embryos or primary mouse embryo fi- sion was higher in tumors in all cases (Supplemental Fig. broblasts (MEFs) lacking Tgifs suggests that the majority S1). Mouse models of intestinal cancer, based on genetic of gene expression changes are unlikely to be due to al- alterations found in human cancers or treatment with tered TGFβ family signaling (Zerlanko et al. 2012; Ander- chemical carcinogens, have been analyzed by gene expres- son et al. 2017). TGIF1 was first identified by its ability to sion array (Kaiser et al. 2007). In these analyses, Tgif1 ex- bind a retinoid response element of the Rbp2 gene and repression was significantly higher in the azoxymethane duce activation by RXR nuclear receptors (Bertolino et al. (AOM) and Apc mutant models but not in one based on in- 1995). TGIFs can bind directly to DNA and repress tran- activation of Smad3, a component of the TGFβ signaling scription via a well-defined consensus site, cTGTCAa, pathway (Supplemental Fig. S2A). Similar results were where the central five bases are most important (Berto- found with Tgif2 expression in this data set, although the lino et al. 1995; Wotton et al. 1999b). Direct repression signal in the AOM samples was too variable to reach stat- via this consensus site has been shown for a small num- istical significance. ber of Tgif target genes (Anderson et al. 2017; Taniguchi To test expression of Tgifs in Apc mutant mouse colon et al. 2017). Recent genome-wide analysis identified a tumors, we combined a Villin-Cre transgene with a loxP- large number of potential Tgif1-binding sites, with en- flanked allele of Apc isolated normal colon and colon tu- richment for the known TGIF consensus element (Lee mors at 12 wk of age and analyzed gene expression by et al. 2015). qRT-PCR. We observed a significant increase in expres- Increased Tgif levels have been implicated in ovarian, sion of both Tgif1 and Tgif2 in colon tumors compared esophageal, and lung cancer among others (Imoto et al. with normal tissue (Fig. 1B). Western blot analysis of sim- 2000; Nakakuki et al. 2002; Wang et al. 2015). Tgif1 pro- ilar 12-wk tumors showed increased Tgif1 protein expres- moted breast cancer progression in a mouse model, inde- sion in tumors compared with normal (Fig. 1C, note that pendent of effects on TGFβ signaling (Zhang et al. 2015). Tgif1 migrates as a doublet due to MAPK mediated phos- The TGIF1 gene was shown to be a direct β-catenin/TCF phorylation; Lo et al. 2001). An increase in Tgif1 expres- transcriptional target that is activated by Wnt/β-catenin sion in small intestine tumors compared with normal signaling, and the possibility that TGIF1 sequesters Axins tissue was also observed (Fig. 1D). We also analyzed tu- to activate Wnt/β-catenin signaling was also suggested as mors in which Tgif1 was deleted specifically from epithe- a mechanism to explain its protumorigenic function lial cells, by including homozygous conditional loxP (Zhang et al. 2015). Recent work with human CRC cell flanked alleles of Tgif1. Little or no Tgif1 signal was lines suggested a role for TGIF1 in CRC progression and detectable in these samples, suggesting that the majority also implicated TGIF1 in controlling the output of the of Tgif1 present in normal colon and its increase in tumors

2 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Tgifs and metabolism

A B of both Tgif1 and Tgif2 is higher in human CRC and in Apc mutant intestinal tumors in mice.

Overexpression of TGIF1 in intestinal epithelium To test effects of increased TGIF1 expression in intestine, we generated a transgene in which T7 epitope-tagged human TGIF1 was expressed from the Villin promoter. Anal- ysis of transgene expression in a panel of tissues by Western blot showed robust expression in the intestine, with no detectable expression in any other tissue examined (Supplemental Fig. S2B). Within the small intestine, we observed readily detectable expression in the proximal, middle, and distal thirds, with much lower expression in the colon and none in nontransgenic tissue (Supplemental Fig. S2C; Parini et al. 2018). To generate mice lacking both C Tgif1 and Tgif2 in the intestinal epithelium, we used Vil- Cre to delete loxP-flanked Tgif1 (as previously described) and loxP-flanked Tgif2, derived from a knockout first allele from EUCOMM. Mice lacking both Tgifs from the intestinal epithelium were viable and grossly normal. Similarly, Vil-TGIF1 transgenic mice were normal and viable to at least 150 d. To compare expression of the Vil-TGIF1 transgene with the endogenous Tgif1 in Apc mutant tumors we performed Western blots with a TGIF1 antiserum that recognizes both human and mouse Tgif1. There was an increase in endogenous Tgif1 expression in regions of the small intestine with tumors, compared with wild-type tissue (Fig. DE 1E). The levels of expression of transgenic TGIF1 were similar in both tumor and normal and, while higher than the expression of mouse Tgif1 in normal tissue, were quite similar to the increased level of endogenous Tgif1 in tumors (Fig. 1E). The transgenic TGIF1 migrates more rapidly on SDS-PAGE than mouse Tgif1, and it appears that expression of the transgene effectively reduces expression of endogenous Tgif1, as evidenced by the almost complete absence of the slower migrating Tgif1 Figure 1. Increased expression of Tgifs in intestine tumors. band in the transgenic samples (Fig. 1E). Thus, Vil- (A) Log2 median centered expression data for TGIF1 and TGIF2 TGIF1 is overexpressed to a level similar to that of the el- in the TCGA colorectal data set (analysis from Oncomine, with evated endogenous expression seen in tumors. upper and lower quartiles and 10th and 90th percentiles), for normal and the indicated tumor types. P-values for comparisons with normal are shown. (B) Relative Tgif1 and Tgif2 expression (mean + Altered tumor burden in the intestine SD of quadruplicate samples) determined by qRT-PCR, from wild- type (normal) colon or from Apc mutant colon tumors. (∗∗) P < To test effects of Tgifs on tumorigenesis, we combined 0.01; (∗∗∗) P < 0.001; (∗∗∗∗) P < 0.0001. (C) Expression of Tgif1 was conditional alleles of Tgif1, both Tgif1 and Tgif2, or the analyzed by Western blot (with Hsp90 as a loading control) from Vil-TGIF1 transgene with Vil-Cre and a heterozygous normal colon (N) and tumor (T) from tissue of the indicated geno- loxP-flanked Apc allele. At 12 wk of age, small intestines types. (+) Wild type; (r) recombined allele. (D) Tgif1 expression were separated into proximal, middle, and distal thirds from the small intestine (normal or tumor, as in C). (E) Expression and opened along the length to identify tumors. Although of Tgif1 in normal (N) and tumor (T) tissue from mice of the indi- the number of tumors per animal was quite variable, there cated genotypes is shown by western blot with a TGIF1-specific was a significant reduction in tumor numbers in mice antiserum and HSP90 as a loading control. Note the transgenic lacking both Tgif1 and Tgif2 and an increase in the Vil- TGIF1 migrates slightly faster than the endogenous mouse Tgif1. TGIF1 mice (Fig. 2A). The number of tumors >1.5 mm in diameter was significantly lower in both the Tgif1 were due to expression in the intestinal epithelium (Fig. and Tgif1;Tgif2 mutants (Fig. 2A). The increase in larger 1C). In support of this, analysis of Tgif1 expression in nor- tumors in the TGIF1-overexpressing mice was highly sig- mal intestinal crypts by immunofluorescence (IF) suggests nificant, whereas there were no significant differences in that Tgif1 is expressed in the majority of epithelial cells the number of smaller (<1.5-mm) tumors. Histological ex- within the crypt (Supplemental Fig. S3). Thus, expression amination of tumors isolated from animals of all four

GENES & DEVELOPMENT 3 Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Shah et al.

Figure 2. Tgif expression promotes intestinal tumorigenesis. (A) The numbers of tumors per animal (at 12 wk) in the small intestine are shown (median, upper, and lower quartiles, 5th and 95th percentiles) for each genotype. Numbers are shown for all tumors and separately for those <1.5 mm or >1.5 mm in diameter. P-values for comparison with the Apc+/r mice are shown. (B) Number of colon tumors per mouse and tumor volume in mm3 are shown (median, upper, and lower quartiles, 5th and 95th percentiles). P-values for comparison with the Apc+/r mice are shown. (∗) P < 0.05; (∗∗) P < 0.01; (∗∗∗) P < 0.001. (C) Representative images of hematoxylin and eosin (H&E)-stained colon tumors from the indicated genotypes are shown. Images were captured at 200× magnification. genotypes revealed no clear differences in tumor morphol- mors, there were no clear histological differences between ogy (Supplemental Fig. S4A). All tumors examined were the colon tumors from mice of each genotype (Fig. 2C). To- adenomas, and we did not observe invasive carcinomas gether, these data suggest that Tgif1 and Tgif2 contribute in these animals. Thus, increasing TGIF1 expression to to Apc mutant intestinal tumorigenesis, and that increas- a level similar to that seen in Apc mutant tumors enhanc- ing TGIF1 expression drives adenoma growth. es adenoma growth, but does not promote transition to invasive adenocarcinoma. Transcriptional changes in Tgif mutant tumors In the middle and distal regions of the small intestine, we observed increased numbers of larger tumors in Vil- To address how increased Tgif levels contribute to intesti- TGIF1 mice and a decrease in total tumor numbers nal tumor growth we performed transcriptome profiling, in Tgif1;Tgif2 mice driven primarily by changes in the comparing normal wild-type colon with colon tumors numbers of larger tumors (Supplemental Fig. S4B,C). Apc from Apc and ATT mice. RNA was isolated from five nor- mutant mice lacking only Tgif1 had an intermediate phe- mal colon samples and from seven tumors from mice of notype between that of the Apc and mice lacking both each of the two genotypes from both males and females. Tgif1 and Tgif2. This was particularly evident when ana- The samples from each of the three genotypes clustered lyzing the proportion of tumors in each mouse that were separately, although there was considerable spread among >1.5 mm in diameter (Supplemental Fig. S4D). Although the tumors, and the two tumor genotypes clustered closer the Vil-Cre;Apc model primarily generates tumors in the to each other than to the wild types (Fig. 3A). To identify small intestine, there are also colon tumors in these ani- genes that were differently expressed, we performed pair- mals. Comparison of tumor number and size in the colon wise comparisons using a 0.5 log2 fold change and an ad- between Apc mice and those lacking Tgif1 did not reveal justed P-value cutoff of <0.01. This identified close to any significant differences (Fig. 2B). However, in the Apc; 2000 genes that were differentially expressed between Tgif1;Tgif2 mice (referred to here as ATT), average tumor the two tumor genotypes, with 884 being higher in the volume was significantly lower, despite the fact that the ATT than in the Apc tumors and 1160 with lower expres- sizes were quite variable (Fig. 2B). Thus, it appears that fur- sion (Supplemental Table S1). Hierarchical clustering of ther reducing overall Tgif levels by deleting Tgif1 and Tgif2 each of these two gene lists suggested that, among the enhances the relatively mild effect of deletion of Tgif1 genes with increased expression in the ATT compared alone, implying redundant function. As with the SI tu- with the Apc tumors, a small fraction was also more

4 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Tgifs and metabolism

A BC

Figure 3. Gene expression changes in Tgif mutant colon tumors. (A) Principle component analysis of RNA sequencing (RNA-seq) data from normal wild-type mouse colon or from colon tumors isolated from Apc heterozygous mice (Apc) or Apc heterozygous mice with homozygous deletion of both Tgif1 and Tgif2 (ATT). Heat maps are shown for all genes with significantly (log2 fold change > 0.5, P-adjusted < 0.01) higher (B) or lower (C) expression in ATT than in Apc. (D) Venn diagrams indicating the overlap between genes that are significantly differently expressed between Apc versus wild-type and Apc versus ATT. (E) Gene set enrichment analysis (GSEA) indicates enrichment of epithelial-to-mesenchymal transition (EMT), KRAS signaling, glycolysis, and glutamine metabolism in Apc tumors compared with ATT. The nominal enrichment score (NES) and false discovery rate q-value are shown. highly expressed in wild-type colon (Fig. 3B). This is con- among genes that are significantly differently expressed in sistent with these genes being Tgif targets that are re- both the Apc to wild type and ATT to Apc comparisons pressed in Apc tumors by increased Tgif expression. (Fig. 3D). However, more genes with higher expression Among the genes with lower expression in ATT tumors, in ATT tumors had reduced expression in Apc tumors many are increased in the Apc tumors compared with compared with wild type (124/392; 31.6%) than had high- wild type, consistent with loss of Tgifs reversing at least er expression (72/392; 18.4%). Thus, it appears that there part of the Apc mutant gene expression program (Fig. is a subset of differentially expressed genes that fit with 3C). Despite the presence of some genes that decrease in being Tgif targets, but there is also significant tumor to tu- the Apc tumor compared with wild type and increase in mor variability and a larger number of genes that do not fit ATT tumors, there was minimal enrichment for this class a simple direct Tgif target model.

GENES & DEVELOPMENT 5 Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Shah et al.

Tgifs are well characterized as repressors of TGFβ-re- clear separation of Apc and ATT tumors, consistent with sponsive transcription (Wotton et al. 1999a; Wotton and the GSEA result, despite the relatively minimal changes Taniguchi 2018) and have been suggested to promote in expression of most components of these pathways Wnt-responsive gene expression (Zhang et al. 2015; (Fig. 4A). Wang et al. 2017). We therefore examined expression of Since Slc2a1 was the most down-regulated glycolytic genes that are known targets of these pathways. For a panel gene in ATT tumors, we examined expression of the of well-characterized TGFβ targets (Smad7, Skil, Serpine1, Glut1 protein in colon tumors of each genotype by IF. Cdkn1a, and Cdkn2b) and some additional genes that re- Glut1 was expressed throughout normal colon and Apc spond to TGFβ in LS1034 CRC cells (Labbé et al. 2007) mutant tumor tissue, with relatively little difference in ex- there was no consistent pattern in expression differences pression between the two (Fig. 4B). In contrast, there was between normal colon and Apc tumors, and deletion of clearly lower expression of Glut1 in the ATT tumor tissue Tgifs had minimal effect (Supplemental Fig. S5A). Over- compared with adjacent normal tissue and with Apc mu- lapping gene expression changes in ATT compared with tant tumors (Fig. 4B). We also examined the RNA sequenc- Apc tumors with expression array data from mouse Apc ing (RNA-seq) data for changes in other metabolic colon tumors or Apc tumors lacking the TGFβ type 2 re- pathways by comparing all genes that were significantly ceptor (GSE82133) (Miguchi et al. 2016) revealed minimal differently expressed between Apc and ATT tumors to overlap (Supplemental Fig. S5B). qRT-PCR analysis of ca- metabolic gene lists from Kyoto Encyclopedia of Genes nonical TGFβ target genes in a set of ATT and Apc colon and Genomes (KEGG). This analysis revealed reduced ex- tumors did not show significant increases in expression pression of multiple genes with links to purine and pyrim- in the Tgif mutants (Supplemental Fig. S5C). idine synthesis and amino acid metabolic pathways (Fig. To examine Wnt signaling we looked at expression of 4C,D). Together, these analyses suggest that loss of Tgifs canonical targets of the pathway. These genes were clear- from Apc tumors results in widespread changes in meta- ly activated in Apc tumors compared with wild-type colon bolic gene expression. but were not further activated by deletion of Tgifs (Supple- mental Fig. S5D). Comparing a set of β-catenin activated Increased expression of acetyl CoA metabolism genes or repressed target genes (Herbst et al. 2014) with our in Tgif mutant tumors data showed limited overlap with expression differences between ATT and Apc tumors, whereas there was clear The majority of metabolic gene expression changes exam- enrichment for these target genes in the comparison be- ined so far are decreases in expression in the absence of tween Apc and wild-type tissue (Supplemental Fig. S5E). Tgifs, suggesting that they are unlikely to be direct Tgif tar- This was supported by qRT-PCR analysis showing in- gets. To identify Tgif target genes we overlapped gene ex- creased expression of Axin2, Lgr5, and Lef1 in Apc mutant pression changes found here with transcriptome profiling tumors, but no decrease in ATT tumors, as would be ex- from wild-type and conditional Tgif1;Tgif2-null (condi- pected if Tgif1 promotes β-catenin-activated gene expres- tional double knockout [cdKO]) mouse embryos (Ander- sion (Supplemental Fig. S5F). Thus, it appears that in the son et al. 2017). There was relatively little overlap context of colon tumors in mice, Tgifs do not play a major between these two data sets, but among the genes that role in regulating either TGFβ or Wnt/β-catenin signaling. changed in both, there was a significant enrichment for genes that increased with deletion of Tgifs from embryos and tumors (Fig. 5A). Wereasonedthatcomparison of these Altered metabolic gene expression in tumors two very different systems might allow us to identify Tgif- lacking Tgifs regulated genes with higher confidence by ruling out sec- To identify functional groups among the gene expression ondary tissue-specific effects. ChIP-seq (chromatin immu- changes, we performed gene set enrichment analysis noprecipitation [ChIP] combined with high-throughput (GSEA). Gene sets indicative of epithelial-to-mesenchy- sequencing) analysis from mouse embryonic stem (ES) mal transition (EMT) and KRAS signaling were among cells identified >12,000 potential Tgif1-bound regions the most significantly enriched in the Apc compared across the genome (Lee et al. 2015). To enrich for higher- with ATT tumors (Fig. 3E). Surprisingly, glycolysis was confidence targets, we considered only the top 40% of pu- also one of the most significantly enriched gene sets in tative Tgif1-bound regions from this analysis, and over- the Apc tumors, and other metabolic signatures were en- lapped this list with genes that were differently expressed riched in Apc compared with ATT tumors (Fig. 3E). Com- in cdKO embryos and tumors lacking Tgifs (Fig. 5B). This paring ATT with Apc tumors, one of the most down- revealed a greater overlap with genes that were activated regulated glycolytic genes was Slc2a1, encoding Glut1, by loss of Tgifs than with genes that had lower expression the major glucose transporter in intestine. For most glyco- in the mutants (Fig. 5B,C). Among the genes with increased lytic enzymes there was a more modest reduction in ex- expression in both cdKO embryos and tumors almost 70% pression in ATT tumors (Fig. 4A). Examining expression had high confidence ChIP peaks (Fig. 5C). of genes encoding proteins that function to generate glu- Analysis of the 125 genes with ChIP-seq peaks and high- cose from pyruvate revealed that these genes were general- er expression in both RNA-seq data sets revealed a signifi- ly slightly more highly expressed in the ATT tumors. cant enrichment for a MEIS1 consensus site (which is Summing the relative expression for each tumor for a panel identical to a TGIF site) associated with these genes, of glycolysis or gluconeogenesis-specific genes revealed a consistent with the idea that they are direct Tgif targets

6 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Tgifs and metabolism

A B

Figure 4. Altered metabolic gene expression in Tgif mutant colon tumors. (A) Heat maps are shown indicating fold-change (comparing ATT with Apc tumors) for the glycolytic pathway and for genes involved specifically in the conversion of pyruvate to glucose. The plot to the right shows summed Z-scores for a panel of genes involved only in glycolysis or in gluconeogenesis, plotted as gluconeogenesis versus glycolysis for each tumor. (B) IF analysis is shown for Glut1 and β-catenin in colon tumors (T) with adjacent normal tissue (N). Heat maps are shown for all genes in the purine and pyrimidine metabolic pathways (C) or amino acid metabolic pathways (D) that are significantly differently expressed comparing Apc with ATT tumors.

(Fig. 5D). Propanoate metabolism and acetyl CoA bio- associated with the early stages of pyruvate metabolism synthetic process were the most significantly enriched (Fig. 4A) and Mpc1 and Pcx expression was increased in pathways, and among the 125 gene list were three genes cdKO embryos, we also examined some changes in this encoding enzymes that synthesize acetyl CoA: Acss1, pathway. Mpc1 expression showed a similar pattern to Acss2, and Mlycd. To place these changes in context, we the acetyl CoA synthetic genes, as did Pcx, although the visualized expression changes for genes encoding a num- increase in Pcx expression in ATT tumors was not statisti- ber of enzymes involved in acetyl CoA metabolism as cally significant (Fig. 5E). This analysis is consistent with part of a metabolic pathway map. Acss2 was significantly the idea that Tgifs directly repress multiple genes involved increased in both cdKO embryos and in tumors lacking in acetyl CoA metabolism and suggests they may also play Tgifs and decreased in Apc tumors compared with wild- a similar function for pyruvate metabolic genes. type colon (Fig. 5E). Similarly, the mitochondrial Acss1 IF analysis of colon tissue from Apc and ATT mice indi- was increased in Tgif mutant embryos and tumors and de- cated that Acss2 expression was reduced in Apc mutant tu- creased in the Apc tumors. Other genes that showed this mors compared with adjacent normal colon, and that pattern included Mlycd, which encodes a cytosolic en- expression was higher in both normal and tumor tissue zyme that converts malonyl CoA to acetyl CoA, and in the ATT mice (Fig. 6A). In both small intestine and Acat1 which generates acetoacetyl CoA from acetyl CoA colon, we observed higher Acss2 expression, with more in the mitochondria as the first step of ketone synthesis evident nuclear localization in the Tgif1;Tgif2 mice com- (Fig. 5E). As there was some increase in expression of genes pared with wild type (Supplemental Fig. S6A,B). In support

GENES & DEVELOPMENT 7 Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Shah et al.

A CE

Figure 5. Identification of putative Tgif target genes. (A) Differentially expressed genes from RNA-seq data from control or Tgif1;Tgif2- null (cdKO) day 9 mouse embryos (GSE78728) overlapped with genes that are significantly differently expressed in ATT versus Apc tumors. (B) Genes with significantly higher (left) or lower (right) expression in either dataset were overlapped with Tgif1 ChIP-seq data from mouse ES cells (GSE55404). (C) The percentage of genes from each of the indicated overlaps between expression data from embryos and tumors with ChIP-seq peaks is shown. (D) EnrichR analysis of the 125 genes with increased expression in embryos and tumors that also have ChIP-seq peaks is shown. (E) A pathway map for selected genes involved in acetyl CoA and pyruvate metabolism is shown, with mitochondrial and cytoplasmic compartments shown separately. Black arrows indicate metabolic reactions, green arrows translocations, and the dashed arrows links to additional metabolic pathways. For each gene shown (boxed), the three colored squares represent fold changes in comparisons of Apc with wild-type (left), ATT with Apc tumor (center), and cdKO with control embryo (right). Larger boxes indicate significant change (P-adjusted < 0.01). Smaller boxes are not significant at this cutoff. of this, Western blot of colon tumors indicated higher showed enrichment for glycolysis in the TGIF1 high and Acss2 expression in ATT than in Apc tumors, and Pcx ex- enrichment for bile acid and fatty acid metabolism in the pression was also higher in the tumors lacking Tgifs (Fig. lowest 10% by TGIF1 expression (Supplemental Fig. 6B). We also analyzed expression of the mitochondrial S7B). Comparison of ACSS2 levels in multiple CRC data Acat1. Expression was clearly higher in ATT tumors sets suggests that levels are reduced in CRC and that lower than in the Apc and in normal small intestine lacking expression correlates with poor survival, a pattern that is both Tgifs (Fig. 6C; Supplemental Fig. S6C). As with opposite to that seen for Tgifs (Supplemental Fig. S7C– Acss2, expression of Acat1 and Mlycd was increased in E). This suggests that similar metabolic gene expression ATT compared with Apc colon tumors (Fig. 6D), consis- differences are seen in human tumors stratified by TGIF1 tent with the idea that loss of Tgifs in colon tumors results expression levels to those seen in our transcriptome profil- in higher expression of both mitochondrial and nuclear/ ing of Apc and ATT tumors. cytoplasmic metabolic regulators. To further explore this potential Tgif function we analyzed human tumor gene ex- Direct Tgif-mediated repression of metabolic gene pression data sets. Separating the human The Cancer Ge- expression nome Atlas (TCGA) CRC data set by TGIF1 expression levels revealed higher expression of ACSS1 and ACSS2 in To address the possibility that genes involved in acetyl the quartile of tumors with lowest TGIF1 expression com- CoA and pyruvate metabolism are direct Tgif targets in pared with the upper quartile (Supplemental Fig. S7A). Fur- multiple cell types, we tested expression of a panel of thermore, GSEA comparing the upper and lower 10% of these genes by qRT-PCR in normal small intestine and tumors from the TCGA PanCancer colon cancer data set in primary MEFs. All three acetyl CoA synthetic genes

8 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Tgifs and metabolism

A B

C D

Figure 6. Increased Acss2 expression in Tgif mutant tumors. (A) IF analysis is shown for Acss2 and β-catenin in colon tumors (T) with adjacent normal (N) tissue from Apc mutants and ATT mice. (B) Western blot analysis of colon tumors from Apc and ATT mice showing expression of Acss2 and Pcx, together with Hsp90 and γ-tubulin loading controls. (C) IF analysis is shown for Acat1 and β-catenin in colon tumors with adjacent normal tissue from Apc and ATT mice. (D) Western blot analysis of Apc and ATT colon tumors showing expression of Acat1 and Mlycd, together with Hsp90 and γ-tubulin loading controls. Molecular weight markers are shown for B and D. and Acat1 were significantly more highly expressed in Acss1 and Acss2 (Fig. 7C). Propionate levels were also Tgif1;Tgif2 null small intestine than in wild-type tissue lower, but did not quite reach significance due to the var- (Fig. 7A). Similarly, expression of Pcx and Mpc1 was also iability among samples. In contrast, other metabolites higher in the mutant. We observed a similar pattern for were essentially unchanged in Tgif mutant small intes- five of the six genes in primary MEFs (Fig. 7B). Acss1 ex- tine compared with the normal (Fig. 7C). Together with pression did not increase in MEFs, but its expression is the gene expression changes seen in both embryos and tu- very low in cultured cells, including primary MEFs. Anal- mors, this is consistent with the possibility that the regu- ysis of metabolite levels in normal small intestine isolated lation of acetate metabolism is a conserved core function from wild-type mice or those lacking both Tgifs suggested of Tgifs. that acetate levels were decreased in the absence of Tgifs We next examined the sequences of the ChIP-seq peaks consistent with increased metabolism by higher levels of associated with each of the six genes analyzed in small

A B C

Figure 7. Tgif1 regulation of metabolic gene expression in normal tissue. (A) Expression of the indicated genes was analyzed by qRT- PCR from normal small intestine from wild-type or cdKO (Tgif1;Tgif2 cdKO) mice. (B) Expression of the same genes was analyzed in wild-type and cdKO primary MEFs. Expression is plotted relative to the wild-type mean (+SD) of four and three replicates for intestine and MEFs. (∗) P < 0.05; (∗∗) P < 0.01; (∗∗∗) P < 0.001, for comparison with wild type. (C) Analysis of metabolites in normal small intestine from wild-type and cdKO mice (mean + SEM). Relative levels of the indicated metabolites are shown with the wild-type level for each set to 1. (∗) P < 0.05 for comparison with wild type.

GENES & DEVELOPMENT 9 Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Shah et al. intestine and primary MEFs. In each case the potential from wild-type small intestine, we observed significant Tgif1-bound region overlapped the transcriptional start enrichment of the putative Tgif1-binding regions from site, and for all but Acss1, at least two TGIF consensus all five genes compared with a negative control region, sites were present (cTGTCA or TGTCAa) (Fig. 8A; Sup- and similar results were obtained from primary MEFs plemental Fig. S8A). To test Tgif1 recruitment, we per- (Fig. 8B,C). We observed significant enrichment for formed ChIP-qPCR for the five genes that had ChIP-seq Acss2 in the wild-type chromatin compared with chroma- peaks with consensus TGIF sites. For Mpc1, we amplified tin from Tgif1;Tgif2 mutant MEFs, indicating the specific- two regions as the predicted peak was quite broad and had ity of this TGIF1 antiserum, as previously demonstrated consensus sites close to each end (Fig. 8A). In chromatin (Supplemental Fig. S8B; Anderson et al. 2017; Taniguchi

A D

Figure 8. Tgif1-mediated repression of Acss2 and Pcx. (A) The positions of the ChIP-seq peaks, qPCR amplicons, and TGIF consensus sites for each to the five genes tested are shown using University of California at Santa Cruz genome browser views. A 4-kb region, centered on the transcriptional start, is shown for each mouse gene, with similarity to human below.(B) Tgif1 binding to each peak region was analyzed by ChIP-qPCR from normal wild-type small intestine. (C) Tgif1 binding in primary MEFs was analyzed by ChIP-qPCR. Chro- matin was precipitated with a TGIF1 antiserum or preimmune serum (pre-I). Data is mean + SD of triplicates and is plotted in arbitrary units with the TGIF1 IP for the negative control region (fbpk3) set equal to 1. (∗) P < 0.05; (∗∗) P < 0.01; (∗∗∗) P < 0.001, for comparison with fbpk3. (D) TGIF1 and TGIF2 were transiently knocked down in HCT116 cells, and expression of the indicated genes analyzed by qRT-PCR. Expression is plotted relative to the nontargeting control knockdown mean (+SD) of 3 replicates. (∗) P < 0.05; (∗∗) P < 0.01, for comparison with control. (E) The Acss2 and Pcx luciferase reporters are shown schematically. Nucleotide positions relative to the transcriptional start site are shown. Black bars are TGIF sites, with the mutant site indicated in white. Gray represents 5′- untranslated se- quence, and white is proximal promoter. (F,G) HCT116 cells were transiently transfected with the indicated luciferase reporters together with increasing amounts of a TGIF1 expression vector (or a control plasmid). Luciferase activity is shown (normalized to a Renilla-luc trans- fectioncontrol), withtheactivityof theAcss2-525-luc(F) andthewild-typePcx-368-luc (G) setto 1. (#)P < 0.01;(##)P < 0.001,forcomparison with no TGIF1 transfection. (∗) P < 0.01; (∗∗) P < 0.001; (∗∗∗) P < 0.0001, for comparison with fold-repression of the wild-type reporter.

10 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Tgifs and metabolism et al. 2017). Transient knockdown of both TGIF1 and a ubiquitin E3 that promotes TGFβ signaling by driving TGIF2 in the human CRC cell line, HCT116, resulted in degradation of the Ski and Skil (SnoN) SMAD corepressors. increased expression of all five metabolic genes, consis- Rnf111 deletion increased tumor numbers in a mouse co- tent with this being a relatively rapid effect of reduced lon cancer model, increased Skil expression, and reduced TGIF levels rather than a longer term response to sus- the TGFβ response (Sharma et al. 2011). Thus, altered tained deletion of Tgif genes (Fig. 8D). To test direct target SMAD corepressor levels can affect CRC tumor progres- gene regulation, we amplified the proximal promoter re- sion, although it remains possible that other Arkadia sub- gions from the Acss2 and Pcx genes and created transcrip- strates contribute. tional reporters. We chose Acss2 because it has a cluster of TGIF1 was proposed to activate Wnt signaling by conserved TGIF consensus sites and might be more sensi- sequestering Axins, allowing activation of Wnt/β-catenin tive to TGIF levels. Pcx, which has only a single conserved target genes (Zhang et al. 2015). In our transcriptome site in the proximal promoter, facilitated the generation of data there is no consistent decrease in expression of canon- a simple mutant version of the reporter in which this site ical Wnt target genes in the absence of Tgifs, as would be was disrupted (Fig. 8E). Since manipulating TGIF levels af- predicted by this model. Although it remains possible fected expression of ACSS2 and PC (human Pcx) in that any effect on Wnt signaling is a cell type-specific func- HCT116 cells, we analyzed Acss2 and Pcx reporter activ- tion of Tgif1, our results suggest the Wnt pathway is not a ity by transient transfection into HCT116. Both the Acss2 major Tgif1 target in colon tumors. It is possible that Tgif1 and Pcx transcriptional reporters were significantly re- promotes β-catenin function, but its effect is masked by pressed by coexpression of increasing amounts of TGIF1 the overriding pathway activation caused by Apc muta- (Fig. 8F,G). Analysis of a deletion mutant of the Accs2 re- tion. However, this still argues against a Tgif effect on tu- porter lacking all TGIF sites and of the Pcx mutant in mor promotion via β-catenin in this mouse model. A which the single site was disrupted revealed significantly further link to Wnt signaling is the demonstration that less repression by TGIF1 than seen with the wild-type re- TGIF1 is directly activated by Wnt/β-catenin signaling porters. In addition, mutant versions of TGIF1, with sin- (Zhang et al. 2015). Our data are consistent with this in gle amino acid changes within the homeodomain that that Tgif1 and Tgif2 expression is increased in Apc mutant prevent DNA binding failed to repress either the Acss2 tumors compared with normal tissue, although we do not or Pcx reporter (Supplemental Fig. S8C). These data sug- know whether Tgif genes are β-catenin targets in the intes- gest that DNA binding by TGIF1 is required to represses tine. Thus, it appears that Tgifs promote intestinal tumor expression of Acss2 and Pcx, and that this repression is growth independent of effects on the two most likely mediated at least in part by conserved consensus TGIF pathways. sites. Taken together, these data suggest that Tgifs are di- Genome-wide analysis of Tgif1 binding to chromatin in rect transcriptional repressors of a set of genes involved in mouse ES cells suggests that a large part of Tgif function is acetyl CoA and pyruvate metabolism. mediated by direct binding to DNA (Lee et al. 2015). In agreement with this, structural studies show that, unlike many other homeodomain proteins, TGIF1 binds with Discussion high specificity and relatively high affinity to its cognate site (Guca et al. 2018). Given the large numbers of gene ex- Using an intestinal cancer model as a starting point to an- pression changes observed in tumors lacking Tgifs we at- alyze TGIF function, we provide evidence that TGIF tran- tempted to identify higher confidence core Tgif target scription factors directly regulate genes involved in acetyl genes by integrating ChIP-seq data with gene expression CoA metabolism. This function of TGIFs does not appear data sets from very divergent systems. Surprisingly, path- to be limited to intestinal tumor or normal tissue and may way analysis of this high confidence gene set identified represent a key unexpected function of these transcription acetyl CoA metabolism as the most significantly enriched factors, independent of the other pathways they are known biological process. In addition, analysis of all gene expres- to regulate. sion changes between Apc tumors and Apc tumors lacking TGIFs are thought to be oncogenic by limiting the anti- both Tgifs revealed changes in multiple metabolic path- proliferative effects of TGFβ signaling. Mutating the ways, further supporting a role for Tgifs as regulators of Tgfbr2 gene in concert with an Apc mutation drives the metabolism. Increased expression of acetyl CoA metabolic transition from adenoma to invasive adenocarcinoma genes is not cell-type specific, occurring in early mouse (Muñoz et al. 2006). Expression of both TGIF1 and embryos, primary MEFs, normal small intestine and colon TGIF2 is higher in CRC than in normal tissue, and Tgif ex- tumors, suggesting that this may be a fundamental, yet un- pression is increased in Apc mutant intestinal tumors. expected, role of Tgifs. We validated a panel of genes with Such increased Tgif levels might be expected to dampen links to acetyl CoA and pyruvate metabolism as direct the TGFβ response and promote tumor growth. Indeed, Tgif1 targets in normal small intestine and primary we show that Tgif expression promotes growth of early ad- MEFs, and demonstrated direct repression of both the enomas. However, transcriptome profiling of colon tu- Acss2- and Pcx-proximal promoters by TGIF1. Full repres- mors revealed almost no overlap with changes in TGFβ- sion was dependent on the presence of conserved TGIF responsive gene expression, suggesting that at least in consensus sites within the promoters, and was abolished this model Tgifs are not major regulators of TGFβ signal- by single amino acid changes in TGIF1 that prevent ing. This is in contrast to the effect of Arkadia (Rnf111), DNA binding. This is consistent with the notion that the

GENES & DEVELOPMENT 11 Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Shah et al. repression of expression of a subset of metabolic genes is a which constitute a major fuel source for colonocytes core Tgif function conserved across multiple cell types (Donohoe et al. 2011). This difference compared with that occurs via direct DNA binding to cognate TGIF sites. most other tissues in the body may impose differing met- This further supports recent work suggesting that while abolic pressures on developing tumors. For example, a re- Tgifs may limit TGFβ-activated gene expression during duction in short chain fatty acid metabolism might embryogenesis, much of the Tgif transcriptional function facilitate the switch to glycolysis. Therefore, it will be of is mediated by direct DNA binding. interest to test functionally whether decreased expression Although we observed extensive changes in expression of ACSS2 or other Tgif targets can promote CRC progres- of metabolic genes, relatively few were increased in the sion. In addition to direct effects of Tgif expression on Tgif mutant tumors, suggesting that the majority of genes involved in acetate metabolism, one of the most dra- changes may be indirect effects. Attempting to place po- matically affected genes encodes the major glucose trans- tential direct Tgif target genes in context indicates that porter, Glut1. High Tgif levels appear to favor Glut1 Tgifs repress anabolic metabolism rather than catabolism. expression specifically in tumors, which may contribute For example, Tgif repression of mitochondrial Acss1 and to a switch away from short chain fatty acid metabolism. Acat1 would be expected to limit ketone synthesis and However, since expression of Slc2a1 increased in the ab- the utilization of acetate by both Acss1 and Acss2 would sence of Tgifs, this is likely not a direct target for Tgif reg- be reduced by Tgif-mediated repression. In support of this, ulation and may be a response to other alterations in we show that in the absence of Tgifs acetate levels are low- metabolism. er in the small intestine, consistent with increased Acss1 In summary, our data suggest a model in which Tgifs and Acss2 activity. However, more extensive analyses of function in multiple cell types to limit expression of a metabolic intermediates is clearly required in both nor- core set of acetyl CoA metabolic genes. In some cancers, mal and tumor tissue to fully validate Tgifs as metabolic including CRC, where Tgif levels increase, this normal regulators. In the absence of Tgifs, anabolic metabolism Tgif function may be co-opted by the tumor as part of may favor synthesis of ketones and sterols, and the utiliza- the metabolic reprogramming. tion of pyruvate to generate other metabolic intermediates. This proposed function of Tgifs is reminiscent of the reprogramming of metabolic gene expression that is Materials and methods recognized as one of the hallmarks of cancer. In addition Mice to an increased reliance on glycolysis, termed the Warburg effect (Warburg 1956), there is extensive rewiring of ener- All animal procedures were approved by the Animal Care and Use gy metabolism in cancer cells (Hanahan and Weinberg Committee of the University of Virginia, which is fully accredit- 2011; Pavlova and Thompson 2016). ed by the AAALAC. Conditional alleles with loxP flanked exons are referred to here as “f” for loxP flanked, or “r” for recombined Recent work suggests that metabolic reprogramming (null). Mice were maintained on a C57BL/6 background. Condi- occurs at the adenoma stage of CRC (Satoh et al. 2017), tional Apc mice were from the NCI (01XAA; B6.Cg-Apctm2Rak/ and it appears that this is downstream from activation of Nci), and the Vilin-Cre line was from Jax [B6.Cg-Tg(Vil1-cre) oncogenes, such as KRAS or BRAF, and requires high 1000Gum/J; 021504] (Madison et al. 2002). Conditional Tgif2 MYC expression (Hutton et al. 2016; Satoh et al. 2017). mice were generated from targeted ES cells from EUCOMM Our data suggest that Tgifs play a role in regulating meta- [Tgif2tm1a(EUCOMM)Wtsi; International Knockout Mouse Consor- bolic gene expression in both normal and tumor tissues tium (IKMC) project 24492; C57BL/6N ES cells] and crossed to and may mediate part of the metabolic reprogramming a conditional Tgif1 line that had been back-crossed for six gener- that occurs in colon adenomas. A role for TGIF1 in regulat- ations to C57BL/6 (Powers et al. 2010). Villin-TGIF1 transgenic ing metabolism in human CRC is supported by GSEA mice were generated at the UVA GEMM Core and were back- crossed to C57BL/6 for at least three generations prior to tumor comparing colon tumors separated by TGIF1 expression analysis. The human TGIF1 cDNA with an amino-terminal T7 levels. Interestingly, one of the best-correlated Tgif target epitope tag was inserted into the Villin promoter plasmid metabolic genes in CRC is ACSS2, which we show is (12.4kbVillin-ΔATG), which was a gift from Deborah Gumucio directly repressed by TGIF1. Several recent studies have (Addgene plasmid 19358; Madison et al. 2002). Germ line trans- linked ACSS2 expression levels to tumor progression mission was verified by PCR and expression by Western blot. (Comerford et al. 2014; Mashimo et al. 2014; Schug et al. 2015). In contrast to our data, high levels of ACSS2 expression correlate with poor prognosis in other cancer types, Tumor analysis, IF, and histology raising the possibility that ACSS2 may function differ- Tissues were fixed in zinc-formalin, paraffin-embedded, sec- ently in CRC. This possibility is supported by analysis of tioned at 5 µm and stained with hematoxylin and eosin (H&E) CRC gene expression data, suggesting that unlike for other or prepared for immunostaining as described (Hao et al. 2018). Im- tumor types, a reduction in ACSS2 expression is observed ages were captured with 10×, 20×, or 40× objectives using a Nikon Eclipse NI-U with a DS-QI1 or DS-Ri1 camera and NIS Elements in CRC and that high expression correlates with better pa- software and adjusted in Adobe Photoshop. For IF, antibodies tient prognosis. were as follows: rabbit anti-Acss2 (Abcam, 66038), mouse One difference that should be noted in this context is the anti-TGIF1 (Santa Cruz Biotechnology, SC-17000), rabbit anti- unusual metabolic environment in which colon tumors Acat1 (Proteintech, 16215-1-AP) rabbit anti-Slc2a1 (Millipore develop. The metabolic activity of gut bacteria results in 07-1401), and mouse anti-β-catenin (BD Transduction Labs, high levels of short chain fatty acids, including acetate, 610153).

12 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Tgifs and metabolism

RNA isolation and qRT-PCR oligonucleotides against human TGIF1 and TGIF2 (as in Ander- son et al. 2017), or with a control siGENOME pool using Dharma- RNA from snap-frozen tissue was isolated and purified using FECT reagent 1. Absolutely RNA kit (Agilent) and quality-checked by Bioana- lyzer. cDNA was generated using SuperScript III (Invitrogen) and analyzed by real-time PCR using a Bio-Rad MyIQ cycler DNA constructs and luciferase assays and Sensimix Plus SYBRgreen plus FITC mix (Bioline) with in- tron-spanning primer pairs selected using Primer3 (http://frodo All luciferase reporter constructs were generated in pGL3 basic .wi.mit.edu). Expression was normalized to Rpl4 and Cyclophilin (Promega) by PCR from genomic DNA. Cells were transfected using the ΔCt method. with firefly luciferase reporters and a phCMVRLuc control (Prom- ega), with pCMV5 TGIF1 as indicated (2, 6, 18 ng per well), using PEI. The two mutant TGIF1 constructs encode TGIF1 with RNA-seq and analysis changes in conserved DNA-binding residues in the homeodo- Poly-A RNA-seq libraries generated with Illumina barcodes were main: either Arg91 altered to methionine (R91M) or Asp88 al- sequenced (NextSeq 500 at the University of Virginia Genome tered to serine (N88S) (Bjerke et al. 2011). After 48 h, activity Analysis and Technology Core) to at least 25 million single-end was assayed with luciferase assay reagent (Biotium) using a Bert- 75-bp reads per sample. Data were analyzed using the Galaxy hold LB953 luminometer. Results were normalized using Renilla server (https://usegalaxy.org). Transcript quantification was per- luciferase activity and assayed with 0.09 μM coelenterazine (Bio- formed using Salmon (Patro et al. 2017) to map to the mm10 synth) as in Hyman-Walsh et al. (2010). Results of replicate trans- mouse genome build, and DESeq2 (Love et al. 2014) within the fections are shown (N = 3; mean + standard deviation) normalized Galaxy site was used for normalizing count data, estimating dis- to the RLuc transfection control. persion, fitting a negative binomial model for each gene, and comparing expression between groups. A cutoff of ±0.5 log2 and an adjusted P-value of <0.01 were considered significant. Enrichment Acknowledgments was analyzed with ENRICHR (http://amp.pharm.mssm.edu/ Enrichr) (Chen et al. 2013; Kuleshov et al. 2016), and heat maps We thank members of the Wotton laboratory for discussions, and were generated with Heatmapper (http://www2.heatmapper.ca/ Dr. M. Mayo for advice. This work was supported by grants from expression) (Babicki et al. 2016). Gene set enrichment was by the National Institute of Neurological Disorders and Stroke GSEA software from the Broad Institute (Mootha et al. 2003; Sub- (NS077958) and the National Institute of General Medical Sci- ramanian et al. 2005). RNA-seq data have been deposited at Gene ences (GM099853) to D.W. Expression Omnibus (GSE116578). Author contributions: A.S., T.A.M., and T.E.F. performed the experiments. A.S., T.A.M., T.E.F., H.F.F., and D.W. analyzed the data. A.S. and D.W. wrote the manuscript. Western blot and metabolite analysis Tissues were ground in PBS followed by addition of NP-40 to 1%, lysates were separated by SDS-PAGE and transferred to Immobi- References lon-P (Millipore), and proteins were visualized using ECL (Pierce). Primary antibodies were against Acss2 (Abcam, 66038), TGIF1 Anderson AE, Taniguchi K, Hao Y, Melhuish TA, Shah A, Turner (Wotton et al. 1999a), Acat1 (Proteintech, 16215-1-AP), Mlycd SD, Sutherland AE, Wotton D. 2017. Tgif1 and Tgif2 repress (Proteintech, 15265-1-AP), γ-tubulin (Sigma, T6557), and HSP90 expression of the RabGAP Evi5l. Mol Cell Biol 37: e00527– (Cell Signaling, 4874). Pcx was detected using Neutravidin-conju- 00516. doi:10.1128/MCB.00527-16 gated HRP (Thermo Fisher). Metabolites from normal small in- Babicki S, Arndt D, Marcu A, Liang Y, Grant JR, Maciejewski A, testine (wild type [N = 4] or lacking Tgif1 and Tgif2 [N = 6]) were Wishart DS. 2016. Heatmapper: web-enabled heat mapping for analyzed by mass spectrometry: Samples were prepared by derivi- all. Nucleic Acids Res 44: W147–W153. doi:10.1093/nar/ tization with 3-NPH and analyzed essentially as described (Han gkw419 et al. 2015). A Waters I-class Acquity chromatography system Bertolino E, Reimund B, Wildt-Perinic D, Clerc R. 1995. A novel inline with a Waters TQS mass spectrometer was used for the sep- homeobox protein which recognizes a TGT core and function- aration and detection. Metabolite levels were normalized to total ally interferes with a retinoid-responsive motif. J Biol Chem protein content in the sample. 270: 31178–31188. doi:10.1074/jbc.270.52.31178 Bjerke GA, Hyman-Walsh C, Wotton D. 2011. Cooperative transcriptional activation by Klf4, Meis2, and Pbx1. Mol Cell Biol ChIP 31: 3723–3733. doi:10.1128/MCB.01456-10 Chromatin was cross-linked for 20 min in 1% formaldehyde and Bürglin TR, Affolter M. 2016. Homeodomain proteins: an update. sonicated to 200–1000 bp using a Branson digital sonifier, with Chromosoma 125: 497–521. doi:10.1007/s00412-015-0543-8 microtip as described (Bjerke et al. 2011). Immunoprecipitation Chen EY, Tan CM, Kou Y, Duan Q, Wang Z, Meirelles GV, Clark was carried out using 10 μL of polyclonal TGIF1 antiserum (Wot- NR, Ma’ayan A. 2013. Enrichr: interactive and collaborative ton et al. 1999a) or preimmune serum. Bound and input fractions HTML5 gene list enrichment analysis tool. BMC Bioinfor- were analyzed by qPCR on a Bio-Rad MyIQ cycler using Sensimix matics 14: 128. doi:10.1186/1471-2105-14-128 Plus SYBR Green plus FITC mix (Bioline). Clevers H, Nusse R. 2012. Wnt/β-catenin signaling and disease. Cell 149: 1192–1205. doi:10.1016/j.cell.2012.05.012 Colnot S, Niwa-Kawakita M, Hamard G, Godard C, Le Plenier S, Cell culture and siRNA knockdown Houbron C, Romagnolo B, Berrebi D, Giovannini M, Perret C. HCT116 cells were cultured in RPMI (Invitrogen) with 10% fetal 2004. Colorectal cancers in a new mouse model of familial ad- bovine serum (Hyclone). Primary MEFs were cultured as de- enomatous polyposis: influence of genetic and environmental scribed (Zerlanko et al. 2012). For knockdown, cells were plated modifiers. Lab Invest 84: 1619–1630. doi:10.1038/labinvest in six-well plates and transfected with Dharmacon SMARTpool .3700180

GENES & DEVELOPMENT 13 Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Shah et al.

Comerford SA, Huang Z, Du X, Wang Y, Cai L, Witkiewicz AK, development by mouse colon tumor models and human colon Walters H, Tantawy MN, Fu A, Manning HC, et al. 2014. Ac- cancer. Genome Biol 8: R131. doi:10.1186/gb-2007-8-7-r131 etate dependence of tumors. Cell 159: 1591–1602. doi:10 Kinzler KW, Vogelstein B. 1996. Lessons from hereditary colorec- .1016/j.cell.2014.11.020 tal cancer. Cell 87: 159–170. doi:10.1016/S0092-8674(00) Donohoe DR, Garge N, Zhang X, Sun W, O’Connell TM, Bunger 81333-1 MK, Bultman SJ. 2011. The microbiome and butyrate regulate Kuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, energy metabolism and autophagy in the mammalian colon. Wang Z, Koplev S, Jenkins SL, Jagodnik KM, Lachmann A, Cell Metab 13: 517–526. doi:10.1016/j.cmet.2011.02.018 et al. 2016. Enrichr: a comprehensive gene set enrichment Feng XH, Derynck R. 2005. Specificity and versatility in TGF-β analysis web server 2016 update. Nucleic Acids Res 44: signaling through Smads. Annu Rev Cell Dev Biol 21: 659– W90–W97. doi:10.1093/nar/gkw377 693. doi:10.1146/annurev.cellbio.21.022404.142018 Labbé E, Lock L, Letamendia A, Gorska AE, Gryfe R, Gallinger S, Fodde R, Edelmann W, Yang K, van Leeuwen C, Carlson C, Moses HL, Attisano L. 2007. Transcriptional cooperation be- Renault B, Breukel C, Alt E, Lipkin M, Khan PM, et al. 1994. tween the transforming growth factor-β and Wnt pathways A targeted chain-termination mutation in the mouse Apc in mammary and intestinal tumorigenesis. Cancer Res 67: gene results in multiple intestinal tumors. Proc Natl Acad 75–84. doi:10.1158/0008-5472.CAN-06-2559 Sci 91: 8969–8973. doi:10.1073/pnas.91.19.8969 Lee BK, Shen W, Lee J, Rhee C, Chung H, Kim KY, Park IH, Kim J. Grady WM, Myeroff LL, Swinler SE, Rajput A, Thiagalingam S, 2015. Tgif1 counterbalances the activity of core pluripotency Lutterbaugh JD, Neumann A, Chang J, Kim S-J, Kinzler KW, factors in mouse embryonic stem cells. Cell Rep 13: 52–60. et al. 1999. Mutational inactivation of transforming growth doi:10.1016/j.celrep.2015.08.067 factor b receptor type II in microsatellite stable colon cancers. Levy L, Hill CS. 2006. Alterations in components of the TGF-β Cancer Res 59: 320–324. superfamily signaling pathways in human cancer. Cytokine Guca E, Suñol D, Ruiz L, Konkol A, Cordero J, Torner C, Aragon Growth Factor Rev 17: 41–58. doi:10.1016/j.cytogfr.2005.09 E, Martin-Malpartida P, Riera A, Macias MJ. 2018. TGIF1 .009 homeodomain interacts with Smad MH1 domain and repress- Lo RS, Wotton D, Massagué J. 2001. Epidermal growth factor sig- es TGF-β signaling. Nucleic Acids Res 46: 9220–9235. doi:10 naling via Ras controls the Smad transcriptional co-repressor .1093/nar/gky680 TGIF. EMBO J 20: 128–136. doi:10.1093/emboj/20.1.128 Han J, Lin K, Sequeira C, Borchers CH. 2015. An isotope-labeled Love MI, Huber W, Anders S. 2014. Moderated estimation of fold chemical derivatization method for the quantitation of short- change and dispersion for RNA-seq data with DESeq2. Ge- chain fatty acids in human feces by liquid chromatography- nome Biol 15: 550. doi:10.1186/s13059-014-0550-8 tandem mass spectrometry. Anal Chim Acta 854: 86–94. Madison BB, Dunbar L, Qiao XT, Braunstein K, Braunstein E, doi:10.1016/j.aca.2014.11.015 Gumucio DL. 2002. cis elements of the villin gene control ex- Hanahan D, Weinberg RA. 2011. Hallmarks of cancer: the next pression in restricted domains of the vertical (crypt) and hori- generation. Cell 144: 646–674. doi:10.1016/j.cell.2011.02.013 zontal (duodenum, cecum) axes of the intestine. J Biol Chem Hao Y, Bjerke GA, Pietrzak K, Melhuish TA, Han Y, Turner SD, 277: 33275–33283. doi:10.1074/jbc.M204935200 Frierson HF Jr, Wotton D. 2018. TGFβ signaling limits lineage Markowitz S, Wang J, Myeroff L, Parsons R, Sun L, Lutterbaugh J, plasticity in prostate cancer. PLoS Genet 14: e1007409. doi:10 Fan RS, Zborowska E, Kinzler KW, Vogelstein B, et al. 1995. .1371/journal.pgen.1007409 Inactivation of the type II TGF-β receptor in colon cancer cells Herbst A, Jurinovic V, Krebs S, Thieme SE, Blum H, Göke B, Kol- with microsatellite instability. Science 268: 1336–1338. ligs FT. 2014. Comprehensive analysis of β-catenin target doi:10.1126/science.7761852 genes in colorectal carcinoma cell lines with deregulated Mashimo T, Pichumani K, Vemireddy V, Hatanpaa KJ, Singh DK, Wnt/β-catenin signaling. BMC Genomics 15: 74. doi:10 Sirasanagandla S, Nannepaga S, Piccirillo SG, Kovacs Z, Foong .1186/1471-2164-15-74 C, et al. 2014. Acetate is a bioenergetic substrate for human Hinoi T, Akyol A, Theisen BK, Ferguson DO, Greenson JK, Wil- glioblastoma and brain metastases. Cell 159: 1603–1614. liams BO, Cho KR, Fearon ER. 2007. Mouse model of colonic doi:10.1016/j.cell.2014.11.025 adenoma-carcinoma progression based on somatic Apc inacti- Massagué J. 2008. TGFβ in cancer. Cell 134: 215–230. doi:10 vation. Cancer Res 67: 9721–9730. doi:10.1158/0008-5472 .1016/j.cell.2008.07.001 .CAN-07-2735 Massagué J, Seoane J, Wotton D. 2005. Smad transcription fac- Hutton JE, Wang X, Zimmerman LJ, Slebos RJ, Trenary IA, Young tors. Genes Dev 19: 2783–2810. doi:10.1101/gad.1350705 JD, Li M, Liebler DC. 2016. Oncogenic KRAS and BRAF drive Melhuish TA, Wotton D. 2000. The interaction of the carboxyl metabolic reprogramming in colorectal cancer. Mol Cell Pro- terminus-binding protein with the Smad corepressor TGIF is teomics 15: 2924–2938. doi:10.1074/mcp.M116.058925 disrupted by a holoprosencephaly mutation in TGIF. J Biol Hyman CA, Bartholin L, Newfeld SJ, Wotton D. 2003. Drosophila Chem 275: 39762–39766. doi:10.1074/jbc.C000416200 TGIF proteins are transcriptional activators. Mol Cell Biol 23: Melhuish TA, Gallo CM, Wotton D. 2001. TGIF2 interacts with 9262–9274. doi:10.1128/MCB.23.24.9262-9274.2003 histone deacetylase 1 and represses transcription. J Biol Chem Hyman-Walsh C, Bjerke GA, Wotton D. 2010. An autoinhibitory 276: 32109–32114. doi:10.1074/jbc.M103377200 effect of the homothorax domain of Meis2. FEBS J 277: 2584– Miguchi M, Hinoi T, Shimomura M, Adachi T, Saito Y, Niitsu H, 2597. doi:10.1111/j.1742-4658.2010.07668.x Kochi M, Sada H, Sotomaru Y, Ikenoue T, et al. 2016. Gasder- Imoto I, Pimkhaokham A, Watanabe T, Saito-Ohara F, Soeda E, min C is upregulated by inactivation of transforming growth Inazawa J. 2000. Amplification and overexpression of TGIF2, factor β receptor type II in the presence of mutated Apc,pro- a novel homeobox gene of the TALE superclass, in ovarian moting colorectal cancer proliferation. PLoS One 11: cancer cell lines. Biochem Biophys Res Commun 276: 264– e0166422. doi:10.1371/journal.pone.0166422 270. doi:10.1006/bbrc.2000.3449 Miller KD, Siegel RL, Lin CC, Mariotto AB, Kramer JL, Rowland Kaiser S, Park YK, Franklin JL, Halberg RB, Yu M, Jessen WJ, Freu- JH, Stein KD, Alteri R, Jemal A. 2016. Cancer treatment and denberg J, Chen X, Haigis K, Jegga AG, et al. 2007. Transcrip- survivorship statistics, 2016. CA Cancer J Clin 66: 271–289. tional recapitulation and subversion of embryonic colon doi:10.3322/caac.21349

14 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

Tgifs and metabolism

Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S, Seo SR, Ferrand N, Faresse N, Prunier C, Abécassis L, Pessah M, Lehar J, Puigserver P, Carlsson E, Ridderstrale M, Laurila E, Bourgeade MF, Atfi A. 2006. Nuclear retention of the tumor et al. 2003. PGC-1α-responsive genes involved in oxidative suppressor cPML by the homeodomain protein TGIF restricts phosphorylation are coordinately downregulated in human di- TGF-β signaling. Mol Cell 23: 547–559. doi:10.1016/j.molcel abetes. Nat Genet 34: 267–273. doi:10.1038/ng1180 .2006.06.018 Moser AR, Pitot HC, Dove WF. 1990. A dominant mutation that Sharma V, Antonacopoulou AG, Tanaka S, Panoutsopoulos AA, predisposes to multiple intestinal neoplasia in the mouse. Sci- Bravou V, Kalofonos HP, Episkopou V. 2011. Enhancement ence 247: 322–324. doi:10.1126/science.2296722 of TGF-β signaling responses by the E3 ubiquitin ligase Arka- Muñoz NM, Upton M, Rojas A, Washington MK, Lin L, Chytil A, dia provides tumor suppression in colorectal cancer. Cancer Sozmen EG, Madison BB, Pozzi A, Moon RT, et al. 2006. Res 71: 6438–6449. doi:10.1158/0008-5472.CAN-11-1645 Transforming growth factor β receptor type II inactivation in- Shibata H, Toyama K, Shioya H, Ito M, Hirota M, Hasegawa S, duces the malignant transformation of intestinal neoplasms Matsumoto H, Takano H, Akiyama T, Toyoshima K, et al. initiated by Apc mutation. Cancer Res 66: 9837–9844. 1997. Rapid colorectal adenoma formation initiated by condi- doi:10.1158/0008-5472.CAN-06-0890 tional targeting of the Apc gene. Science 278: 120–123. doi:10 Nakakuki K, Imoto I, Pimkhaokham A, Fukuda Y, Shimada Y, .1126/science.278.5335.120 Imamura M, Amagasa T, Inazawa J. 2002. Novel targets for Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, the 18p11.3 amplification frequently observed in esophageal Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander squamous cell carcinomas. Carcinogenesis 23: 19–24. doi:10 ES, et al. 2005. Gene set enrichment analysis: a knowledge- .1093/carcin/23.1.19 based approach for interpreting genome-wide expression pro- Oshima M, Oshima H, Kitagawa K, Kobayashi M, Itakura C, files. Proc Natl Acad Sci 102: 15545–15550. doi:10.1073/ Taketo M. 1995. Loss of Apc heterozygosity and abnormal tis- pnas.0506580102 sue building in nascent intestinal polyps in mice carrying a Taniguchi K, Anderson AE, Sutherland AE, Wotton D. 2012. Loss truncated Apc gene. Proc Natl Acad Sci 92: 4482–4486. of Tgif function causes holoprosencephaly by disrupting the doi:10.1073/pnas.92.10.4482 SHH signaling pathway. PLoS Genet 8: e1002524. doi:10 Parini P, Melhuish TA, Wotton D, Larsson L, Ahmed O, Eriksson .1371/journal.pgen.1002524 M, Pramfalk C. 2018. Overexpression of transforming growth Taniguchi K, Anderson AE, Melhuish TA, Carlton AL, Man- factor β induced factor homeobox1 represses NPC1L1 and ukyan A, Sutherland AE, Wotton D. 2017. Genetic and molec- lowers markers of intestinal cholesterol absorption. Athero- ular analyses indicate independent effects of TGIFs on Nodal sclerosis 275: 246–255. doi:10.1016/j.atherosclerosis.2018.06 and Gli3 in neural tube patterning. Eur J Hum Genet 25: 208– .867 215. doi:10.1038/ejhg.2016.164 Patro R, Duggal G, Love MI, Irizarry RA, Kingsford C. 2017. Salm- Vogelstein B, Fearon ER, Hamilton SR, Kern SE, Preisinger on provides fast and bias-aware quantification of transcript ex- AC, Leppert M, Nakamura Y, White R, Smits AM, Bos JL. pression. Nat Methods 14: 417–419. doi:10.1038/nmeth.4197 1988. Genetic alterations during colorectal-tumor develop- Pavlova NN, Thompson CB. 2016. The emerging hallmarks of ment. N Engl J Med 319: 525–532. doi:10.1056/NEJM198 cancer metabolism. Cell Metab 23: 27–47. doi:10.1016/j 809013190901 .cmet.2015.12.006 Wang Y, Wang H, Gao H, Xu B, Zhai W, Li J, Zhang C. 2015. Ele- Polakis P. 1995. Mutations in the APC gene and their implica- vated expression of TGIF is involved in lung carcinogenesis. tions for protein structure and function. Curr Opin Genet Tumour Biol 36: 9223–9231. doi:10.1007/s13277-015-3615-8 Dev 5: 66–71. doi:10.1016/S0959-437X(95)90055-1 Wang JL, Qi Z, Li YH, Zhao HM, Chen YG, Fu W. 2017. TGFβ in- Powers SE, Taniguchi K, Yen W, Melhuish TA, Shen J, Walsh CA, duced factor homeobox 1 promotes colorectal cancer develop- Sutherland AE, Wotton D. 2010. Tgif1 and Tgif2 regulate Nod- ment through activating Wnt/β-catenin signaling. Oncotarget al signaling and are required for gastrulation. Development 8: 70214–70225. doi:10.18632/oncotarget.19603 137: 249–259. doi:10.1242/dev.040782 Warburg O. 1956. On the origin of cancer cells. Science 123: 309– Sancho E, Batlle E, Clevers H. 2004. Signaling pathways in intes- 314. doi:10.1126/science.123.3191.309 tinal development and cancer. Annu Rev Cell Dev Biol 20: Wotton D, Taniguchi K. 2018. Functions of TGIF homeodomain 695–723. doi:10.1146/annurev.cellbio.20.010403.092805 proteins and their roles in normal brain development and hol- Satoh K, Yachida S, Sugimoto M, Oshima M, Nakagawa T, Aka- oprosencephaly. Am J Med Genet C Semin Med Genet 178: moto S, Tabata S, Saitoh K, Kato K, Sato S, et al. 2017. Global 128–139. doi:10.1002/ajmg.c.31612 metabolic reprogramming of colorectal cancer occurs at ade- Wotton D, Lo RS, Lee S, Massagué J. 1999a. A Smad transcription- noma stage and is induced by MYC. Proc Natl Acad Sci 114: al corepressor. Cell 97: 29–39. doi:10.1016/S0092-8674(00) E7697–E7706. doi:10.1073/pnas.1710366114 80712-6 Schug ZT, Peck B, Jones DT, Zhang Q, Grosskurth S, Alam IS, Wotton D, Lo RS, Swaby LA, Massagué J. 1999b. Multiple modes Goodwin LM, Smethurst E, Mason S, Blyth K, et al. 2015. Ace- of repression by the smad transcriptional corepressor TGIF. J tyl-CoA synthetase 2 promotes acetate utilization and main- Biol Chem 274: 37105–37110. doi:10.1074/jbc.274.52.37105 tains cancer cell growth under metabolic stress. Cancer cell Zerlanko BJ, Bartholin L, Melhuish TA, Wotton D. 2012. Prema- 27: 57–71. doi:10.1016/j.ccell.2014.12.002 ture senescence and increased TGFβ signaling in the absence Segditsas S, Tomlinson I. 2006. Colorectal cancer and genetic al- of Tgif1. PLoS One 7: e35460. doi:10.1371/journal.pone terations in the Wnt pathway. Oncogene 25: 7531–7537. .0035460 doi:10.1038/sj.onc.1210059 Zhang MZ, Ferrigno O, Wang Z, Ohnishi M, Prunier C, Levy L, Seo SR, Lallemand F, Ferrand N, Pessah M, L’Hoste S, Camonis J, Razzaque M, Horne WC, Romero D, Tzivion G, et al. 2015. Atfi A. 2004. The novel E3 ubiquitin ligase Tiul1 associates TGIF governs a feed-forward network that empowers Wnt sig- with TGIF to target Smad2 for degradation. EMBO J 23: naling to drive mammary tumorigenesis. Cancer Cell 27: 547– 3780–3792. doi:10.1038/sj.emboj.7600398 560. doi:10.1016/j.ccell.2015.03.002

GENES & DEVELOPMENT 15 Downloaded from genesdev.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press

TGIF transcription factors repress acetyl CoA metabolic gene expression and promote intestinal tumor growth

Anant Shah, Tiffany A. Melhuish, Todd E. Fox, et al.

Genes Dev. published online February 26, 2019 Access the most recent version at doi:10.1101/gad.320127.118

Supplemental http://genesdev.cshlp.org/content/suppl/2019/02/26/gad.320127.118.DC1 Material

Published online February 26, 2019 in advance of the full issue.

Creative This article is distributed exclusively by Cold Spring Harbor Laboratory Press for the first Commons six months after the full-issue publication date (see License http://genesdev.cshlp.org/site/misc/terms.xhtml). After six months, it is available under a Creative Commons License (Attribution-NonCommercial 4.0 International), as described at http://creativecommons.org/licenses/by-nc/4.0/. Email Alerting Receive free email alerts when new articles cite this article - sign up in the box at the top Service right corner of the article or click here.